Text-to-Image
Diffusers
English
learn_ddpm / model /model.py
Harshit Agarwal
slides added
ee90412
'''
THis gile is to contain the DDPM implementation modularized for loading, prediciton and training.
'''
from torch import nn
import math
import torch
from utils import forward_diffusion_sample, sample_timestep, sample_plot_image
import torch.nn.functional as F
from attn_utils import SelfAttention, CBAM, Block_CBAM
class Block(nn.Module):
def __init__(self, in_ch, out_ch, time_emb_dim, up=False):
super().__init__()
self.time_mlp = nn.Linear(time_emb_dim, out_ch)
if up:
## up channel - gobig big big bigg from smol smol smol with 3x3 kernel
self.conv1 = nn.Conv2d(2*in_ch, out_ch, 3, padding=1)
self.transform = nn.ConvTranspose2d(out_ch, out_ch, 4, 2, 1)
else:
self.conv1 = nn.Conv2d(in_ch, out_ch, 3, padding=1)
self.transform = nn.Conv2d(out_ch, out_ch, 4,2,1)
self.conv2 = nn.Conv2d(out_ch, out_ch, 3, padding=1)
self.relu = nn.ReLU()
self.batch_norm1 = nn.BatchNorm2d(out_ch)
self.batch_norm2 = nn.BatchNorm2d(out_ch)
def forward(self, x, t, ):
h = self.batch_norm1(self.relu(self.conv1(x)))
time_emb = self.relu(self.time_mlp(t))
time_emb = time_emb[(..., ) + (None, ) * 2]
h = h + time_emb
h = self.batch_norm2(self.relu(self.conv2(h)))
return self.transform(h)
class PositionEmbeddings(nn.Module):
def __init__(self,dim):
super().__init__()
self.dim = dim
def forward(self, time):
device = time.device
half_dim = self.dim // 2
embeddings = math.log(10000) / (half_dim - 1)
embeddings = torch.exp(torch.arange(half_dim, device=device) * -embeddings)
embeddings = time[:, None] * embeddings[None, :]
embeddings = torch.cat((embeddings.sin(), embeddings.cos()), dim=-1)
return embeddings
class SimpleUnet(nn.Module):
def __init__(self):
super().__init__()
image_channels = 3
down_channels = (64, 128, 256, 512, 1024)
up_channels = (1024, 512, 256, 128, 64)
self.device = "cuda" if torch.cuda.is_available() else "cpu"
out_dim = 3
time_emb_dim = 32
## timestep stored as positional encoding in terms of sine
self.time_mlp = nn.Sequential(
PositionEmbeddings(time_emb_dim),
nn.Linear(time_emb_dim, time_emb_dim),
nn.ReLU()
)
self.conv0 = nn.Conv2d(image_channels, down_channels[0], 3, padding=1)
self.down_blocks = nn.ModuleList([
Block(down_channels[i], down_channels[i+1], time_emb_dim)
for i in range(len(down_channels)-1)
])
self.up_blocks = nn.ModuleList([
Block(up_channels[i], up_channels[i+1], time_emb_dim, up=True)
for i in range(len(up_channels)-1)
])
## readout layer
self.output = nn.Conv2d(up_channels[-1], out_dim, 1)
def forward(self, x, timestep):
t = self.time_mlp(timestep)
x = self.conv0(x)
residual_inputs = []
for down in self.down_blocks:
x = down(x, t)
residual_inputs.append(x)
for up in self.up_blocks:
residual_x = residual_inputs.pop()
x = torch.cat((x, residual_x), dim=1)
x = up(x, t)
return self.output(x)
@torch.no_grad()
def sample(self, noise):
"""
Generate an image by denoising a given noise tensor using the reverse diffusion process.
Args:
noise (torch.Tensor): Initial noise tensor (e.g., sampled from a Gaussian distribution).
Returns:
torch.Tensor: Denoised image.
"""
img = noise # Start with the provided noise tensor
T = self.num_timesteps # Total timesteps for diffusion
stepsize = 1 # You can adjust if needed
# Iterate through the timesteps in reverse order
for i in range(0, T)[::-1]:
t = torch.full((noise.size(0),), i, device=noise.device, dtype=torch.long) # Current timestep
img = sample_timestep(self, img, t) # Perform one reverse diffusion step
img = torch.clamp(img, -1.0, 1.0) # Clamp the image to ensure values stay in [-1, 1]
return img
def get_loss(self, x_0, t):
x_noisy, noise = forward_diffusion_sample(x_0, t, self.device)
noise_pred = self(x_noisy, t)
return F.l1_loss(noise, noise_pred)
def train(self, dataloader, BATCH_SIZE=64,T=300, EPOCHS=50, verbose=True):
from torch.optim import Adam
device = "cuda" if torch.cuda.is_available() else "cpu"
self.to(device)
optimizer = Adam(self.parameters(), lr=0.001)
epochs = EPOCHS
for epoch in range(epochs):
for step, batch in enumerate(dataloader):
optimizer.zero_grad()
t = torch.randint(0, T, (BATCH_SIZE,), device=device).long()
loss = self.get_loss(self, batch[0], t)
loss.backward()
optimizer.step()
if verbose:
if epoch % 5 == 0 and step % 150 == 0:
print(f"Epoch {epoch} | step {step:03d} Loss: {loss.item()} ")
sample_plot_image(self)
def test():
## TODO: add the testing loop here
pass
################################################################################################
####################### ATTENTION LAYERS ADDEDD TO THE MODEL ###################################
################################################################################################
class SimpleUnetWSelfAttn(nn.Module):
def __init__(self):
super().__init__()
image_channels = 3
down_channels = (64, 128, 256, 512, 1024)
up_channels = (1024, 512, 256, 128, 64)
out_dim = 3
time_emb_dim = 32
## timestep stored as positional encoding in terms of sine
self.time_mlp = nn.Sequential(
PositionEmbeddings(time_emb_dim),
nn.Linear(time_emb_dim, time_emb_dim),
nn.ReLU()
)
self.num_timesteps = 300
self.conv0 = nn.Conv2d(image_channels, down_channels[0], 3, padding=1)
self.down_blocks = nn.ModuleList([
Block(down_channels[i], down_channels[i+1], time_emb_dim)
for i in range(len(down_channels)-1)
])
self.up_blocks = nn.ModuleList([
Block(up_channels[i], up_channels[i+1], time_emb_dim, up=True)
for i in range(len(up_channels)-1)
])
self.self_attention = SelfAttention(down_channels[-1])
## readout layer
self.output = nn.Conv2d(up_channels[-1], out_dim, 1)
# def settimestep()
def forward(self, x, timestep):
self.num_timesteps = timestep
t = self.time_mlp(timestep)
x = self.conv0(x)
residual_inputs = []
for down in self.down_blocks:
x = down(x, t)
residual_inputs.append(x)
x = self.self_attention(x)
for up in self.up_blocks:
residual_x = residual_inputs.pop()
x = torch.cat((x, residual_x), dim=1)
x = up(x, t)
return self.output(x)
@torch.no_grad()
def sample(self, noise):
"""
Generate an image by denoising a given noise tensor using the reverse diffusion process.
Args:
noise (torch.Tensor): Initial noise tensor (e.g., sampled from a Gaussian distribution).
Returns:
torch.Tensor: Denoised image.
"""
img = noise # Start with the provided noise tensor
T = self.num_timesteps # Total timesteps for diffusion
stepsize = 1 # You can adjust if needed
print(noise.device)
# Iterate through the timesteps in reverse order
for i in range(T - 1, -1, -1):
t = torch.full((noise.size(0),), i, device=noise.device, dtype=torch.long) # Current timestep
img = sample_timestep(self, img, t) # Perform one reverse diffusion step
img = torch.clamp(img, -1.0, 1.0) # Clamp the image to ensure values stay in [-1, 1]
return img
################################################################################################
#################### Convolutional Block Attention Module ADDED TO THE MODEL ###################
################################################################################################
class SimpleUnetWCBAM(nn.Module):
def __init__(self):
super().__init__()
image_channels = 3
down_channels = (64, 128, 256, 512, 1024)
up_channels = (1024, 512, 256, 128, 64)
out_dim = 3
time_emb_dim = 32
## timestep stored as positional encoding in terms of sine
self.time_mlp = nn.Sequential(
PositionEmbeddings(time_emb_dim),
nn.Linear(time_emb_dim, time_emb_dim),
nn.ReLU()
)
self.num_timesteps = 300
self.conv0 = nn.Conv2d(image_channels, down_channels[0], 3, padding=1)
self.down_blocks = nn.ModuleList([
Block_CBAM(down_channels[i], down_channels[i+1], time_emb_dim)
for i in range(len(down_channels)-1)
])
self.up_blocks = nn.ModuleList([
Block_CBAM(up_channels[i], up_channels[i+1], time_emb_dim, up=True)
for i in range(len(up_channels)-1)
])
self.self_attention = SelfAttention(down_channels[-1])
## readout layer
self.output = nn.Conv2d(up_channels[-1], out_dim, 1)
# def settimestep()
def forward(self, x, timestep):
self.num_timesteps = timestep
t = self.time_mlp(timestep)
x = self.conv0(x)
residual_inputs = []
for down in self.down_blocks:
x = down(x, t)
residual_inputs.append(x)
x = self.self_attention(x)
for up in self.up_blocks:
residual_x = residual_inputs.pop()
x = torch.cat((x, residual_x), dim=1)
x = up(x, t)
return self.output(x)
@torch.no_grad()
def sample(self, noise):
"""
Generate an image by denoising a given noise tensor using the reverse diffusion process.
Args:
noise (torch.Tensor): Initial noise tensor (e.g., sampled from a Gaussian distribution).
Returns:
torch.Tensor: Denoised image.
"""
img = noise # Start with the provided noise tensor
T = self.num_timesteps # Total timesteps for diffusion
stepsize = 1 # You can adjust if needed
print(noise.device)
# Iterate through the timesteps in reverse order
for i in range(T - 1, -1, -1):
t = torch.full((noise.size(0),), i, device=noise.device, dtype=torch.long) # Current timestep
img = sample_timestep(self, img, t) # Perform one reverse diffusion step
img = torch.clamp(img, -1.0, 1.0) # Clamp the image to ensure values stay in [-1, 1]
return img